PMLE Monitoring ML Solutions Practice Question

A team uses Vertex AI Explainable AI with integrated gradients for a deep learning model. They want to reduce the computational cost of explanations without significantly reducing explanation quality. Which configuration change should they make?

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Reduce the number of integral approximation steps.

Integrated gradients approximates Shapley values by integrating gradients along a path. Reducing the number of steps (integral approximation steps) reduces computation, but may reduce quality. A moderate reduction balances cost and quality.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Switch from integrated gradients to XRAI.
Why it's wrong here
XRAI is for images; may not be applicable to deep learning models in general.
✓
Reduce the number of integral approximation steps.
Why this is correct
Fewer steps lower computation; optimal steps can be tuned.
✗
Apply feature attribution to only a random subset of predictions.
Why it's wrong here
This would reduce coverage, not computational cost per request.
✗
Use sampled Shapley instead, as it is always cheaper.
Why it's wrong here
Sampled Shapley can be cheaper but may not preserve quality equally; also not the only option.

Go deeper

PMLE Monitoring ML Solutions Practice Question

Why each option matters

Option-by-option breakdown

Related to this question